The long-range goal of the EarLab project is to create realistic, web-based models capable of predicting human auditory responses to a wide range of acoustic stimuli or environmental insults. Applications range from predicting how acoustic over-stimulation leads to hearing impairment to explaining how humans are able to function in complex acoustic environments. To achieve our goal we will synthesize and integrate existing information on mammalian hearing and create tools that will facilitate effective interaction between investigators from different research disciplines. The models will improve our ability to take the knowledge obtained from animal models and apply it to humans. In the process of constructing and integrating the EarLab models and tools, data gaps will be identified that will need to be filled by future animal and/or human studies. The neuroscience objectives are to: 1) develop an integrated model of cochlear function, 2) develop a model of noise-induced hearing loss, and 3) develop integrated models of sound-source localization. The informatics goals are: 1) to create the computational infrastructure for a web-based auditory modeling environment, 2) to create an intelligent virtual laboratory that will allow rapid reconfiguration of simulations and interchange of model modules, as well as reference information to aid simulation design, and 3) to develop a user interface for model configuration and data visualization.